After shutdown/restart, ath10k sometimes stops receiving packets

Kalle Valo kvalo at
Mon Jun 16 02:30:07 PDT 2014

Avery Pennarun <apenwarr at> writes:

> On Sun, Jun 15, 2014 at 1:45 AM, Kalle Valo <kvalo at> wrote:
>> Avery Pennarun <apenwarr at> writes:
>>> Someone in our QA group seems to be able to produce it after two or
>>> three test runs with Veriwave.  Before each run, he stops and restarts
>>> hostapd.
>> That's quite often. Was he testing DFS or something else? Maybe this is
>> related to radar detection or channel switching?
> He's not using DFS channels.

Good, so we can rule out that.

> This is supposedly just a simple throughput test. (Although I've
> noticed in the past that the Veriwave sometimes produces odd sequences
> of packets.)

Yeah, these kind of test equipments can behave different than normal
802.11 devices. For example, I remember one very expensive test box was
acting as an AP and it just ignored PM bit from the clients. No wonder
the packet loss was high :)

In this case it might that the way Veriwave behaves so which makes it
easy to reproduce the bug. And that's a good thing.

Any chance you could get full trace-cmd logs (ath10k, mac80211, cfg80211
and hostapd) when the bug fails with Veriwave? I'm sure that would get
more hints what's happening.

Kalle Valo

More information about the ath10k mailing list